Skip to content

When recording, what are audio artifacts and dithering noise?

January 2, 2011

A computer never stores a perfect representation of sound, only an increasingly accurate representation. The number of bits in a sample (8 bit, 16 bit, 24 bit, etc), say how accurately the quiet parts of a recorded sound can be stored. With 8 bits, only the loudest parts can be stored accurately. When the quiet parts are stored inaccurately, you don’t get silence, you get a kind of electronic static. Every bit that you add to the sample doubles the accuracy, or, said another way, each bit added means that the static is about half the volume as before the bit was added. At 12 bits, the accuracy is good enough to where it isn’t loud and annoying, but you can faintly hear it in the background. At 16 bits, the static is extremely quiet.

If it is so quiet at 16 bits, then it makes since to wonder why people record at 24 bit. The static is always there, regardless of what you record. If you record a quiet instrument, then the static is faint, but it is at the same level as it would have been if you’d recorded a brass ensemble or amplified guitar. If, however, you recorded that quiet sound, but later need to turn it up, you can do so, but, while you turn up the quiet sound, you are also turning up the static. Turning it up a little will still mean that the static is fairly quiet. If, though, you need to process it in a way that dramatically increases the level, such as putting the sound through an amp sim plug in, then that static will also be dramatically amplified. Recording at 24 bit gives you room to turn up a sound, or otherwise boost it as part of the mix, and not have to worry about bringing up the level of the static, also.

Recording at 24 bit means that the static is as low as possible in the original recording. DAW software will commonly use a higher resolution of internal processing, such as 32 or 64 bits, in order to avoid introducing any additional digital static in to the recorded signal as it is amplified, compressed, and otherwise processed as part of the mix.

If you’ve listened to both cassettes and CDs, you’ll remember that cassettes have a background hiss that is always there. The hiss from tapes is similar to what happens when sampling at low bit resolutions. Tapes don’t store sound in the same way that happens with a computer, but the noise of a typical cassette is similar to the noise level of a sample with 10 or 12 bit resolution. CDs, on the other hand, have 16 bit resolution. Using a CD instead of a cassette, to avoid the background hiss of the cassette, is a similar situation to how people use higher sample rates on a computer to avoid the low level static of digital sampling.

From → Ask Bryan

  1. Glenn permalink

    I still don’t know what an artifact is. lol

  2. Amy permalink

    agree with Glenn

  3. It makes since …..

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s